JP7408389B2 - Method for controlling proliferation of eukaryotic cells - Google Patents

Description

FIELD OF THE INVENTION The present invention relates to the field of biotechnology, and more particularly to the field of controlling the growth of cells, particularly eukaryotic cells.

BACKGROUND AND RELATED ART Proteins of therapeutic or commercial importance are often produced by culturing mammalian cells containing a nucleic acid encoding a recombinant protein of interest. For example, a seed train process is used to generate sufficient numbers of mammalian cells to inoculate the mass production bioreactors that are actually used to produce and recover the protein of interest. A conventional seed train process starts with a small cell sample, such as a cryopreserved cell bank vial, followed by multiple cell culture expansion steps in progressively larger culture vessels or tanks.

Batch and fed-batch production processes consist of a nonproductive growth phase in which the cell population accumulates, and a more productive quiescent phase in which the majority of the protein or other biological product is produced inside or outside the cell. Including period. Although proliferative and stationary phases can also be observed in prokaryotic cells, these two distinct phases are important because biological products produced by eukaryotic cells are affected by the metabolic state of the producing cells (e.g., the glycosylation pattern of protein products). of particular relevance for the culture of eukaryotic cells, as they are particularly sensitive to

Cell growth in cell culture is a process that is very sensitive to many different factors: small differences in the composition of the culture medium or nutrient solution, small differences in the cell density of the cell sample used to inoculate the bioreactor. Differences, or inaccurate calibrations of various devices measuring process parameters such as pH, temperature, oxygen or carbon dioxide concentrations, can lead to differences in the rate of cell growth, in terms of the amount of protein produced by each cell, and even in Significant differences may result in the chemical composition or modification of the product. Proteins produced by eukaryotic cells undergo many different biochemical modification processes, such as glycosylation and phosphorylation. Any change in process parameters and growth conditions can have a significant impact on cellular metabolism and thus the glycosylation pattern of the produced proteins. For example, Gammell P, Barron N, Kumar N, Clynes M (2007) “Initial identification of low temperature and culture stage induction of miRNA expression in suspension CHO-K1 cells”. J Biotechnol 130:213-218 (Non-Patent Document 1) identified miR-21, a growth-inhibitory miRNA, to be upregulated during stationary phase growth induced by a temperature shift to lower temperatures or during normal batch culture. Therefore, changing the temperature in an ongoing eukaryotic cell culture project can affect cell conditions.

US 2009/0104594 A1 describes a bioreactor comprising a sensor coupled to a model-free adaptive controller or optimizer. The sensor can provide real-time measurements of quality that correlate with final product titer or other desired product quality characteristics. Additionally, methods for culturing living cells are described. The method includes the steps of incubating cells in a bath, measuring multiple conditions inside the bath, and individually comparing the multiple measurements to multiple set points using a model-free adaptive controller. and adjusting conditions inside the bath based on the at least one comparison.

Pharmaceutical companies therefore try to keep process parameters as constant as possible with respect to the culture of eukaryotic cells to ensure reproducible and safe production of the desired biological product within a defined time interval. There is.

Eukaryotic cells are often grown under two different temperature conditions for a proliferative phase and a stationary phase: the proliferative phase is typically carried out at 37°C, whereas the productive phase is typically To ensure a standardized and reproducible production process, standardized product concentration and quality at the end of the production process, reduce the temperature to approximately 34℃ and maintain the temperature constant at 34℃ throughout the production process. It is carried out by Similarly, the temperature is maintained constant during the growth phase to ensure reproducibility of the process.

Therefore, state-of-the-art approaches to culturing eukaryotic cells are based on maintaining constant culture parameters, especially temperature (at least throughout each of the proliferative and stationary phases).

However, slight variations in the initial cell density, medium composition, and calibration characteristics of the measurement equipment cannot be completely avoided, and therefore the duration of the cell culture project and the biological products that can be recovered after a given period of time cannot be completely avoided. There still exists variability regarding the total amount of , which makes it very difficult for biotech and pharmaceutical companies to plan their production processes.

US 2009/0104594 A1

Gammell P, Barron N, Kumar N, Clynes M (2007) “Initial identification of low temperature and culture stage induction of miRNA expression in suspension CHO-K1 cells” J Biotechnol 130:213-218

Summary The object of the invention is to provide an improved method and system for controlling the proliferation of cells, as specified in the independent claims. Aspects of the invention are given in the independent claims. Aspects of the invention may be freely combined with each other if they are not mutually exclusive.

In one aspect, the invention relates to a method for controlling proliferation of eukaryotic cells. Control logic receives the target cell density and target time. Target cell density represents the desired cell density of eukaryotic cells at the target time. The target time represents a future point in time. Eukaryotic cells are cultured in the culture medium in the first tank. The method includes: for each of the plurality of time intervals (e.g., at the beginning or end of each time interval);
- measuring the cell density of eukaryotic cells in the culture medium;
- calculating, by means of control logic, a predicted cell density at a target time, at least as a function of the measured cell density;
- Comparing the predicted cell density and the target cell density by means of control logic;
- by means of control logic, adjusting the temperature of the culture medium in the first tank if the predicted cell density deviates from the target cell density; Adjustment is by increasing the temperature by 0.1°C to 1°C if the predicted cell density is lower than the target cell density; or decreasing the temperature by 0.1°C to 1°C if the predicted cell density is higher than the target cell density. It is carried out either by.

Said feature may be advantageous as it allows the proliferation of eukaryotic cells to be actively controlled to yield a defined number of cells at a defined time point. Surprisingly, although the temperature was not kept constant during cell culture growth, for many different cell culture projects even when the starting conditions (starting temperature, starting cell concentration) were slightly different, the target It was observed that the target cell density was reached in time. Eukaryotic cells and their metabolism are very sensitive to changes in environmental temperature, but at least temperature regulation does not exceed a temperature difference of 1 °C per predicted time interval and neither exceeds nor falls below the physiologically permissible temperature range. It has been observed that the biological products produced do not differ from those produced under constant temperature conditions.

Furthermore, it was observed that the cells were able to tolerate and rapidly recover from temperature changes within the specified temperature ranges above. Therefore, by adjusting the temperature of the medium within the specified temperature range above, the overall time of the cell culture project will not be extended. This differs for adjustment of other production parameters, for example in the case of pH, some addition is required to recover from the pH change before the cells can accumulate the cell population and continue producing the desired biological product. pH was observed to reduce the efficiency of cell production as it was observed that it required a long time.

By repeatedly predicting the time at which the currently growing cell culture will likely reach the desired cell density, comparing the predicted time to the specified target time, and adapting the cell culture temperature accordingly, Embodiments of the present invention allow for compensating for slight variations in cell culture starting conditions, such that the cell culture reaches the desired cell density at a certain defined future time (the "target time") and achieves the desired It may be possible to ensure that quantities of biological products are produced. Depending on the cell line, culture medium, conditions in the tank, and type of biological product produced, cell culture starts from inoculation of the cell culture until the time the cell culture reaches the target cell density. The typical duration of projects varies. Many projects have a typical duration in the range of 3-5 days, for example 4.5 days. It has been observed that embodiments of the invention can compensate for deviations of up to 20% in one or more starting process parameters from the "typical" or "reference" process parameters of a reference cell culture project. For example, if a reference project is run at a constant temperature of 37°C and takes exactly 4 days (96 hours) to reach the desired cell density, our method using a dynamic prediction-based temperature adaptation approach Embodiments may make it possible to achieve the same cell density within a 20% shorter time interval, ie within 3.2 days (77 hours). Similarly, if the amount of cells used to inoculate a bioreactor is 20% less than the amount of cells used to inoculate a reference bioreactor, dynamic temperature adaptation according to embodiments of the invention will reduce the number of starting cells. to provide the desired cell density at the desired target time.

Similarly, embodiments of the present invention eliminate errors (failures in oxygen or nutrient supplies, failures in pH control systems, failures in oxygen or nutrient supply, failures in pH control systems, , etc.) was observed to successfully compensate for several causes of

Accordingly, aspects of the present invention may enable compensating for system failures and deviations in process parameters relative to a reference culture project, and thus enable better control and predictability of cell culture projects.

Calculation of the predicted cell density at the target time, at least as a function of the measured cell density, may be performed based only on the currently measured cell density. Preferentially, the cell density at the target time is predicted as a function of multiple measured cell densities measured in previous time intervals.

According to a preferred embodiment, the temperature adjustment does not reduce the temperature of the culture below a temperature of 32°C and does not increase the temperature of the culture to a temperature of approximately 38°C.

It may be beneficial to carry out temperature regulation only within said temperature range, since temperature regulation within said range may influence the growth rate in a controlled manner, but not in the direction of other metabolic pathways. , for example, was observed not to significantly shift cellular metabolism in the direction of the production of heat shock proteins. Temperature adjustment within this range was observed to be completely and rapidly reversible, as the cells do not require extra time to recover from the previously used temperature.

According to embodiments, the method for controlling cell proliferation, and optionally temperature regulation, is performed throughout the growth phase, or throughout the stationary phase, or throughout the growth and quiescence time.

The above may be beneficial because the longer the period used for dynamic adjustment of cell culture temperature, the better the ability to compensate for system component failures and suboptimal starting conditions.

According to embodiments, the first tank includes an instrument for performing on-line measurements of cell density in the culture medium. Therefore, the cell density of eukaryotic cells measured at each of the plurality of time intervals is an online measurement.

For example, measurement devices that measure the dielectric properties of a culture medium to determine the number of cells in a given volume of medium, or optical techniques, for example in the form of an online microscope, for automatic counting of living cells in a medium. It can be used for.

For example, the FOGALE biomass system can be used to determine cell density in a tank without taking samples. The use of dielectric sensors, unlike optical techniques, has the advantage that the system is not sensitive to air bubbles, cell debris, and other particles in the suspension. By using a FOGALE biomass sensor or similar sensor device, viable biomass concentration (biovolume) can be determined as an on-line measurement independent of cell size variations.

Another example of an online cell densitometer is an in situ microscope. In situ microscopy is a system developed to acquire images of mammalian cells directly (in situ) inside a bioreactor during cell culture projects. It requires only minimal operator intervention. Examples of in situ microscopy and its uses can be found in Joeris K1, Frerichs JG, Konstantinov K, Scheper T: "In-situ microscopy: Online process monitoring of mammalian cell cultures" Cytotechnology. 2002 Jan; 38(1-3):129- 34. doi: 10.1023/A:1021170502775.

A further example of an online cell density meter is a Raman spectrometer. The use of Raman spectroscopy to determine cell density in bioreactors is described, for example, by Justin Moretto et al. in "Process Raman Spectroscopy for In-Line CHO Cell Culture Monitoring" (April 2011 issue of American Pharmaceutical Review - Volume 14 ) is described by.

Performing online measurements of cell density may have the advantage of avoiding the problems associated with offline measurements: due to the significant latency between the time of measurement and the time of performing each control operation, offline Measurement data tends to be of low quality. Furthermore, the sampling process for obtaining measurements from the sample can increase the risk of infection of the cell culture by undesirable bacteria and reduce the accuracy of the measurements due to sampling effects (e.g. changes in temperature). there is a possibility.

In a further beneficial aspect, the use of online cell density rather than offline density allows for fully automated cell culture growth control. Furthermore, using online cell density can provide a large number of cell density measurements during a cell culture project, thus providing a better data base for predicting cell density at a target time, This can also improve prediction accuracy.

According to an aspect, the control logic is configured to terminate at least the adjustment of the temperature when a termination criterion is met. Termination criteria may be, for example, a determination that a target time has been reached or that the measured cell density is equal to or higher than the target cell density.

Optionally, the control logic may not only stop regulating the temperature, but also stop growing the cell culture. For example, the control logic can send one or more control commands to an automated control element that causes the control element to stop feeding and aerating cells in a cell culture tank, and to automate a cell harvesting process. start. Alternatively, the control logic sends a message to one or more of the appropriate technical staff's mobile communication devices to notify the staff that the cells in the tank must be retrieved or further processed. Additionally or alternatively, the messages are displayed on a screen operably coupled to the control logic. Further processing may involve transferring the cells, or the entire culture containing the cells, from the first tank to a different, preferentially larger tank, such as another pre-production or production bioreactor. Depending on the embodiment, the collection and/or transfer steps may be performed manually, automatically, or semi-automatically.

Preferentially, the iterative prediction of the future cell density at the target time, the comparison with the target time, and the adjustment of the temperature if necessary, are performed completely automatically. According to a preferred embodiment, the repeated cell transfers from one pre-production tank to another pre-production tank or production tank and the final recovery of the cells or the biological product produced are also completely automatic, i.e. It is carried out without any human intervention. Robust to small variations in starting conditions, robust to failure of system components during the cell culture phase, and capable of producing defined cell densities and defined products at specified future times without any human intervention. This can be advantageous as it can provide a fully automated system and method that can reliably deliver quality.

According to an aspect, control logic receives a cell concentration excursion threshold for adjusting temperature. The control logic adapts the temperature of the culture medium in the first tank only if the expected cell concentration at the target time deviates from the target cell concentration by more than the threshold. This is important because the threshold introduces some inertia within the control loop and ensures that the temperature is adapted only when deviations from the target cell concentration are expected to reach a significant level. can be beneficial. For example, the control logic receives a lower cell concentration excursion threshold, e.g., 95% of the target cell density, and controls the temperature of the medium in the first tank only if the predicted cell concentration is lower than 95% of the target cell concentration. It can be raised. Additionally or alternatively, the control logic receives an upper cell concentration excursion threshold, e.g., 105% of the target cell density, and controls the first tank only if the predicted cell concentration is higher than 105% of the target cell concentration. can lower the temperature of the medium within. Other lower and upper thresholds can be used as well, such as 99% and 101%, or 98% and 102%, or a lower cell density threshold of 99% and an upper cell density threshold of 105%. such as asymmetric thresholds.

According to aspects, the control logic provides a graphical user interface (GUI). The GUI allows the user to enter the target time and target cell density before the control logic begins calculating the predicted cell density. The GUI allows the user to modify the target time and/or target cell density while the control logic calculates the predicted cell density at multiple time intervals.

According to some embodiments, the method further includes transferring the culture medium in the first tank and the eukaryotic cells contained therein from the first tank to the second tank after reaching the target cell density. The second tank is larger than the first tank. Control logic receives additional target cell densities and additional target times. The additional target cell density represents the desired cell density of eukaryotic cells at the additional target time. Further growth control steps in the second tank are carried out after the cells in the first tank have reached the target cell density, so the further target time is later than the time the cells were transferred to the second tank. represents time.

For example, the control logic can receive information about target cell densities, further target cell densities, target and/or additional target times. Alternatively, the control logic may read the (further) target time and target cell density from a configuration file created or modified by the user before the control logic is instantiated.

The transferred eukaryotic cells are then cultured in a second tank. For each of the plurality of further time intervals (e.g., at the beginning or end of each time interval), the following sub-method is performed:
- measuring the cell density of eukaryotic cells in the culture medium of the second tank;
- calculating, by control logic, further predicted cell densities at further target times, at least as a function of the measured cell densities;
- comparing the further predicted cell density with the further target cell density by control logic;
- adjusting the temperature of the culture medium in the second tank if the further predicted cell density deviates from the further target cell density by means of control logic; Adjustments are performed by increasing the temperature by 0.1°C to 1°C if the further predicted cell density is lower than the further target cell density. If the additional predicted cell density is higher than the additional target cell density, the control logic reduces the temperature by 0.1°C to 1°C.

Depending on the embodiment, the first target cell density and the second target cell density may be the same or different from each other.

According to some embodiments, the first tank and the second tank are each a bioreactor configured to grow the cell culture as a batch culture. The method is used in a seed train culture process to generate a defined minimum number of cells for inoculating a production bioreactor.

The above may be advantageous because the cell culture can be grown semi-automatically or fully automatically to a first target cell density at a first target time. The cell culture can then be transferred to another larger tank, such as a further pre-production bioreactor or tank. The cells are then grown to a second target cell density at a further second target time. The steps can be repeated for three, four, or even more pre-production vessels or bioreactors until a sufficient number of cells are obtained to inoculate the large-scale production bioreactor. Thus, the method can be used to propagate cell cultures in seed train culture projects.

According to embodiments, the first tank is a bioreactor configured to grow the cell culture as a pre-production batch culture. The second tank is a bioreactor that is larger than the first tank and configured to grow the cell culture as a production culture.

Production culture may be, for example, fed-batch culture. In some embodiments, the production culture may also be a batch culture.

According to an embodiment, the eukaryotic cell is a mammalian cell, in particular a Chinese hamster ovary (CHO) cell. Alternatively, the eukaryotic cell is a baby hamster kidney (BHK) cell or a human cell line. An example of a human cell line is Per.C6, a cell derived by immortalizing human embryonic retinal cells with the adenovirus E1 gene. This is often used to produce monoclonal antibodies (MAbs) using G418 as the selection agent. Additionally, the eukaryotic cell may be an NS0 cell, a Sp2/0 cell, a COS cell, a K562 cell, or a HEK cell.

All embodiments described herein can similarly be used to control the growth of prokaryotic cells, particularly bacterial cells, such as Escherichia coli cells.

According to embodiments, the culture medium in the first tank has a volume of 500 liters or less.

In accordance with an aspect, the method includes, for each of the plurality of time intervals (e.g., at the beginning or end of each time interval):
- defining for the current time interval a current observation interval having a predetermined length and ending at the end of the current time interval; and receiving all cell densities measured during the current observation interval; The method further includes the step of:

The predicted cell density is calculated as a function of at least the measured cell density. The calculation involves extrapolating the cell density measured during the current observation interval up to the target time.

For example, the control logic may define the current observation interval by determining the current time provided by a clock and parsing a configuration file or user input to determine the length of the observation interval. The control logic then identifies the current observation time interval as a time interval that ends at the current time and begins the determined length before the current time. An example of observation intervals determined at different current times is described in FIG. 2.

According to embodiments, extrapolation involves performing a curve fitting operation or regression analysis on the cell density measured during the current observation time.

According to an aspect, all time intervals of the plurality of time intervals are of equal length.

According to embodiments, all time intervals of the plurality of time intervals are of different lengths.

According to an aspect, the method includes determining whether the current time is close to the target time. For example, a) a predetermined duration of time has elapsed since inoculation of the culture, or b) a predetermined number of predictions of cell density at the target time have already been performed, or c) a predetermined percentage of the target cell density. or d) a predetermined time before the target time, such as 13 hours before the target time (typically, the predetermined time before the target time ranges from 12 to 14 hours). ), the control logic may determine that the current time is close to the target time. Other methods exist to automatically determine whether the current time is close to the target time. Alternatively, the user can enter an indication via the GUI that the current time is close to the target time.

In response to determining that the current time is close to the target time, the control logic shortens the length of all future time intervals of the plurality of time intervals.

The above can be beneficial, as the accuracy of cell density prediction and the accuracy of growth control can be improved: at the end of the culture time, the cell density is already high, and a small error in the prediction and a long time between each prediction The combination of time intervals may not result in exactly the desired cell density at the target time. To increase the accuracy of prediction and growth control, embodiments of the invention use shorter prediction time intervals at the end of a culture project. For example, shortening the predicted time interval may be performed completely automatically by the control logic as described above, or manually by the user.

According to embodiments, each of the plurality of time intervals has a duration of less than 30 minutes, preferentially less than 10 minutes. For example, each of the time intervals has a duration within the range of 1 minute to 10 minutes.

According to an embodiment, the predetermined length of the observation interval is in the range from 60 minutes to 480 minutes, especially from 80 to 100 minutes, such as 90 minutes.

It was observed that the above durations of prediction time interval and observation interval provide a good compromise between high prediction accuracy and resource consumption (comparison of measurement acquisition, prediction and predicted cell density is (which can consume processing power and generate network traffic; updating the screen containing the curves showing measured and predicted cell densities can further consume network, memory, and CPU capacity).

According to an embodiment, the culture medium in the first tank (and optionally also the second tank and other tanks) is CD-CHO AGT medium. However, the appropriate medium is highly dependent on the cell type, the phase of cell growth (e.g., proliferative or stationary phase), the type of product to be produced, the type of bath or bioreactor used, and other factors. . Therefore, other media such as DMEM, RPMI, Hams F12, and others may also be used.

According to a further embodiment, adjusting the temperature comprises increasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C, or decreasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C.

According to embodiments, the GUI or another component of the control system allows a user to enter additional control parameters before and/or during execution of a cell culture project. For example, the user can determine the length of the prediction time interval, the length of the observation interval, the length of the adjustment interval, upper and/or lower thresholds for considering the predicted cell density to be (sufficiently) lower or higher than the target cell density. , and/or the amount of temperature adjustment per adjustment action. Preferentially, the user is able to enter different values for the positive and negative temperature regulation phases. This may be a better adaptation of temperature control to the heating and cooling characteristics of the bioreactor.

In a further aspect, the invention relates to a method for providing a desired cell density of eukaryotic cells at a defined future time. The method includes inputting a desired cell density and future time by a human user via an interface of control logic. Desired cell density represents the desired cell density of eukaryotic cells at the input future time. According to the method of any one of the embodiments described herein, control logic is used to control proliferation of eukaryotic cells. Growth control is performed such that the entered desired cell density is used as the target cell density, the entered future time is used as the target time, and the eukaryotic cells are provided at the target cell density at the target time. be done.

In a further aspect, the invention relates to a system for controlling eukaryotic cell proliferation. The system should be controlled to produce a defined number of cells at a defined future time. The system includes control logic and a processor configured to execute the control logic. Control logic is operably coupled to a first tank containing eukaryotic cells in culture. The control logic is configured to receive a target cell density and a target time. For example, target cell density and target time may be received from a network interface, read from a configuration file of control logic, or received as user input via a graphical user interface. Target cell density represents the desired cell density of eukaryotic cells at the target time. The target time represents a future point in time.

For each of the plurality of time intervals (e.g., at the beginning or end of each time interval), the system:
- receiving a measured cell density of eukaryotic cells in the culture medium of the first tank;
- calculating the predicted cell density at the target time as a function of at least the measured cell density;
- comparing the predicted cell density with the target cell density;
- adjusting the temperature of the culture medium in the first tank when the predicted cell density deviates from the target cell density according to the control logic, and if the predicted cell density is lower than the target cell density; The control logic is configured to perform a sub-method comprising increasing the temperature by 0.1° C. to 1° C. and, if the predicted cell density is higher than the target cell density, the control logic decreasing the temperature by 0.1° C. to 1° C. Ru.

According to embodiments, the system further comprises a first tank, and optionally also one or more further tanks (when the cells reach a desired cell density in the previously used tank, the further tank is added to the first tank). or can receive cell culture from another tank).

According to embodiments, the first tank includes an instrument for performing on-line measurements of cell density in the culture medium. The cell density of eukaryotic cells measured at each of the plurality of time intervals is an online measurement.

According to an aspect, the control logic is configured to terminate at least the adjustment of the temperature when a termination criterion is met. Termination criteria may be, for example, that the target time has been reached or that the measured cell density is equal to or greater than the target cell density.

In accordance with embodiments, the system or one of its components, such as control logic, is configured to provide a graphical user interface (GUI). The GUI allows the user to enter the target time and target cell density before the control logic begins calculating the predicted cell density. The GUI may also allow a user to modify the target time and/or target cell density while the control logic calculates the predicted cell density at multiple time intervals.

According to embodiments, the system includes at least a first tank and a second tank. The first tank is adapted to allow a user to transfer the culture medium in the first tank and the eukaryotic cells contained therein from the first tank to the second tank. For example, the first tank may include an opening or pipe for manually or automatically transferring the medium containing the cells from the first tank to the second tank.

The second tank is larger than the first tank and is adapted to receive the cell-containing medium and additional cell-free medium, as well as to grow the transferred eukaryotic cells.

The control logic is configured to receive additional target cell densities and additional target times, for example by reading additional target times from a storage medium or by receiving target times from a user via a GUI or network interface. . The additional target cell density represents the desired cell density of eukaryotic cells at the additional target time. Additional target times represent future time points (after the time the cells were transferred from the first tank to the second tank).

The cell densitometer in the second tank measures the number of eukaryotic cells in the culture medium in the second tank for each of a plurality of time intervals, referred to herein as "further time intervals" or "further predicted time intervals." configured to measure density. The control logic is configured such that for each further time interval,
- calculating further predicted cell densities at further target times, at least as a function of said measured cell densities;
- Comparing further predicted cell densities with further target cell densities;
- configured to adjust the temperature of the culture medium in the second tank if the further predicted cell density deviates from the further target cell density; The adjustment is to increase the temperature by 0.1°C to 1°C if the further predicted cell density is lower than the further target cell density; or to increase the temperature by 0.1°C if the further predicted cell density is higher than the further target cell density. Including lowering by ~1°C.

According to embodiments, the first tank is a bioreactor configured to grow the cell culture as a batch culture. The second tank is a bioreactor that is larger than the first tank and configured to grow the cell culture as a batch culture. The first tank and the second tank are adapted for use in a seed train culture process to generate a defined minimum number of cells for inoculating a production bioreactor.

According to embodiments, the second tank includes an instrument for measuring cell density in the second tank and includes a temperature control device for adjusting the temperature of the medium in the second tank according to temperature control commands of the control logic. .

According to embodiments, the first tank is a bioreactor configured to grow the cell culture as a pre-production batch culture. The second tank is a bioreactor that is larger than the first tank and configured to grow the cell culture as a production culture.

According to an embodiment, the eukaryotic cell is a mammalian cell, in particular a Chinese hamster ovary (CHO) cell.

According to embodiments, the culture medium in the first tank has a volume of 500 L or less.

According to an aspect, the control logic, for each of the plurality of time intervals,
- define for the current time interval a current observation interval that has a predetermined length and ends at the end of the current time interval; and - defines all cell densities measured during the current observation interval. and configured to perform receiving.

Calculating the predicted cell density as a function of at least the measured cell density includes extrapolating the cell density measured during the current observation interval up to the target time.

According to embodiments, extrapolation involves performing a curve fitting operation or regression analysis on the cell density measured during the current observation time.

According to embodiments, the time intervals are of equal length.

Depending on the embodiment, the time intervals are of different lengths. In particular, at least some time intervals in the last third of the cell culture project are shorter than time intervals in the first third of the cell culture project.

According to some aspects, control logic is configured to determine whether the current time is close to the target time. In response to determining that the current time is close to the target time, the control logic shortens the length of all future time intervals of the plurality of time intervals.

According to an embodiment, each of the plurality of time intervals has a duration in the range of less than 30 minutes, preferentially less than 10 minutes, such as in the range of 1 to 10 minutes.

According to embodiments, the predetermined length of the observation interval is within the range of 60 minutes to 480 minutes.

According to embodiments, adjusting the temperature comprises increasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C, or decreasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C.

According to an aspect, the control logic is configured to perform a temperature adjustment of the first tank only if at least an adjustment time interval has elapsed since a previous temperature adjustment of the first tank by the control logic. The duration of the adjustment time interval is within the range of 60-120 minutes.

According to embodiments, the control logic allows a human user (e.g., by providing a GUI) to input a desired cell density and a future time to reach the desired cell density through an interface of the control logic. It is composed of Desired cell density represents the desired cell density of eukaryotic cells at the input future time. The control logic is further configured to control proliferation of eukaryotic cells according to any one of the embodiments described herein. Growth control is performed such that the entered desired cell density is used as the target cell density, the entered future time is used as the target time, and the eukaryotic cells are provided at the target cell density at the target time. be done.

A "tank" as used herein is a container for holding, transporting, or storing a liquid. The tank may be, for example, a bioreactor, in particular a pre-production or production bioreactor, a vial, or a recovery or transport tank. For example, a tank has contents suitable for culturing a plurality of cells (e.g., recombinant mammalian cells) in liquid culture under a set of controlled physical conditions that allow maintenance or proliferation of the cells. It may be a container with a volume. A non-limiting example of a tank is a bioreactor (eg, any of the exemplary bioreactors described herein or known in the art).

As used herein, a "bioreactor" is a vessel in which a chemical process involving an organism or a biochemically active substance derived from such an organism is carried out. This process may be, for example, aerobic or anaerobic. A number of different bioreactor types exist, differing in shape (eg, cylindrical or otherwise), size (eg, milliliters, liters to cubic meters), and materials (stainless steel, glass, plastic, etc.). According to embodiments, the bioreactor is adapted to grow cells or tissues in cell culture. Depending on the embodiment and/or mode of operation, the bioreactor may be a batch bioreactor, a fed-batch bioreactor, or a continuous bioreactor (eg, a continuous stirred tank reactor model). An example of a continuous bioreactor is a chemostat.

As used herein, "control logic" refers to a piece of program logic, e.g., an application program or module, a computer chip, or a tank containing a cell culture, configured to receive and process one or more measurements from a tank containing a cell culture. configured to process the measurements as well as to send one or more control commands to the tank or a hardware module operably coupled to the tank to control growth of the cell culture. Another piece of software, hardware, or firmware, or a combination thereof. For example, the control logic may be a program module that is part of or interoperates with the bioreactor monitoring and control software. The control logic may also be part of a laboratory information management system (LIMS).

A “measuring device” or “instrument” that is “operably connected” to a tank is, for example, permanently or temporarily located inside the tank or connected to a wall of the tank and included in or on the tank. The measuring device may be a measuring device configured to measure one or more physical or chemical parameters of a cell culture or culture medium. Measurements may be performed automatically in response to commands or periodically.

As used herein, "on-line measurement" is the process of obtaining measurements that characterize the condition of a tank or the cell culture or medium contained therein, over a period of time necessary to perform the measurement. is a process shorter than the time for which the characteristic changes significantly. A significant change may be a change that exceeds a predetermined threshold. For example, a change of greater than 5% may be considered a significant change. The threshold may vary depending on different characteristics. Online measurements may make it possible to control the bioreactor in real time. In particular, the "on-line measurements" used are carried out directly, i.e. directly on the tank or on the cell culture or medium contained therein, without taking a sample from the culture medium and measuring said sample. It may be a measurement.

"Off-line measurement" is the process of obtaining measurements that describe a characteristic of the condition of a tank or the cell culture or medium contained therein, in which the period of time required to perform the measurement is such that the characteristic is significant. It is a process that takes longer than the time that can change. A significant change may be a change that exceeds a predetermined threshold. Typical examples of off-line measurements are automatic, semi-automatic or manual sampling of the medium, eg to determine the current cell density. Off-line measurements are based on discrete sampling steps. Off-line measurements are not recommended due to the significant latency between the time of measurement and the time each control operation is performed, since the bioreactor characteristics may have changed since the sample was taken. Data-driven bioreactor controls tend to be of low quality. A significant change may be a change above a predetermined threshold, such as 2% or any other percentage value, depending on each state characteristic.

"Raising or lowering the temperature by X °C" means, by way of example, raising or lowering the temperature of an object relative to the measured current temperature at the same time that a prediction of cell density is carried out, for example. . According to embodiments, the tank includes an on-line thermometer.

As used herein, the process of "curve fitting" is the process of constructing a curve or mathematical function that best fits a set of data points (in this case: measured cell density), possibly subject to constraints. Curve fitting may involve either an exact fit to the data or a smoothing in which a "smooth" function is constructed that approximately fits the data.

As used herein, "regression analysis" is a process that further focuses on problems of statistical inference, such as how much uncertainty exists in the curve fit to the observed data along with random errors. The fitted curve is optionally displayed on the GUI. Fitted curves are used to infer the values of functions for which data are not available and to summarize relationships between two or more variables.

As used herein, the term "regression analysis" refers to a statistical process for estimating a relationship between variables and representing that relationship as a function, such as a linear or nonlinear curve function. Most commonly, regression analysis estimates the conditional expected value of a dependent variable given the independent variables, that is, the mean value of the dependent variable if the independent variables were held fixed. For example, current measured cell density and a growth model, such as a linear or logarithmic growth model, can be used to estimate the relationship between future time and predicted cell density. Many techniques have been developed for performing regression analysis. For example, linear regression or ordinary least squares regression can be used. Performing a regression analysis may involve estimating a continuous response variable, as opposed to a discrete response variable used in classification (metric regression).

As used herein, the term "extrapolation" refers to the use of observed data to calculate predicted data values that extend beyond the range of the observed data (e.g., by fitting or interpolating observed data). (by) using a curve derived from observed data. For example, extrapolation may predict future data values by fitting a curve through observed data values obtained in past time intervals. The curve may be a linear or non-linear curve. The range of observation data may be, for example, historical time intervals during which cell density was measured within a particular tank. The time interval is also referred to herein as the "observation interval." Since the observation time interval is determined anew for each prediction, it is also referred to as the "current observation interval." Calculations of future curve sections are subject to some degree of uncertainty because they may reflect the method used to construct the curve as much as they reflect observed data.

Eukaryotic cells may be cells derived from any human and non-human mammal, such as humans, hamsters, mice, green monkeys, rats, pigs, cows, or rabbits, among others. For example, the mammalian cell may be an immortalized cell. In some embodiments, the mammalian cell is a differentiated cell. In some embodiments, the mammalian cell is an undifferentiated cell. Additional examples of mammalian cells are known in the art.

As used herein, the term "seed train step" means that the number of starting cells (e.g., recombinant mammalian cells) in the first cell culture is typically at an initial cell density of greater than 0.25 x ¹⁰ cells/mL. It is a multi-step method that is scaled up to N-1 cell cultures containing a sufficient number of cells to inoculate a production bioreactor. A seed train tank is therefore a vessel or bioreactor adapted for use in a seed train process. The purpose of the seed train is to generate a sufficient number of cells to inoculate the production bioreactor. From the thawing volume used to maintain the cell line, cells are transferred to a number of culture systems that grow larger with each passage (e.g., T-flasks, roller bottles or shake flasks, small-scale bioreactor systems, and subsequently larger bioreactors). By passage, the number of cells must be increased. A single-use system can be applied, inoculating the system partially full and then increasing the culture volume by addition of medium). Typically, production bioreactors are inoculated from the largest seed drain scale. Seed trains provide space for optimization, such as choosing the optimal time point for passage of cells from one scale to a larger scale. According to some embodiments, the cells in each tank used in the seed train process are grown to a cell density that is the same as the target cell density of the production bioreactor within a defined period of time. In each of the seed train tanks and production tanks, growth control methods according to embodiments described herein can be applied based on the same common target cell density (but typically at each seed train stage). Depending on process specificity, target times will vary. In other embodiments, different stages in the seed train process use different target cell densities.

As used herein, a "batch culture" is a cell culture in which growth occurs in tanks that are operated in a batch mode. If the tank, e.g. a bioreactor, is operated in batch mode, the medium remains the same throughout the entire cultivation period. Therefore, the culture medium is not added or replenished (as in a flow process), nor is it partially replaced (as in a chemostat bioreactor). In batch culture, cell proliferation, defined as the increase in cell number in batch culture, typically increases very strongly (log phase) after an initial homeostasis (lag phase), then gradually decreases (transition phase). / “stationary phase”), and becomes negative if necessary (decline phase). The reason for growth stagnation in batch culture is the depletion of the nutrient medium and the accumulation of toxic substances.

As used herein, a "fed-batch culture" is a cell culture in which growth occurs in tanks operated in a fed-batch mode. The tank may in particular be a production tank containing a plurality of cells in a liquid culture. In fed-batch mode, fresh liquid culture medium is periodically or continuously added to the tank without substantial or significant removal of liquid culture medium from the tank during cultivation. The fresh culture medium may be the same culture medium that was present in the tank at the beginning of the culture. In some instances of fed-batch culture, the fresh liquid culture is a concentrated form of the liquid culture that was present in the tank at the beginning of the culture.

As used herein, a "graphical user interface (GUI)" is a type of user interface that allows a user to interact with an electronic device, such as a computer or a smartphone, through graphical icons and visual indicators. A GUI allows the user, through direct manipulation of graphical elements of the GUI (e.g. icons, buttons, bars, text areas, etc.), to implement software and/or hardware-based functions, e.g. control logic or hardware modules of a tank (e.g. thermostat).

The term "culture" or "cell culture" refers to the maintenance or growth of mammalian cells (eg, recombinant mammalian cells) under a controlled set of physical conditions. The term "culture of mammalian cells" or "cell culture" means a liquid culture containing a plurality of mammalian cells that is maintained or grown under a set of controlled physical conditions. do.

The term "liquid culture medium" or "culture medium" means a liquid containing sufficient nutrients to allow cells (eg, mammalian cells or bacteria) to grow or multiply in vitro. For example, liquid culture solutions include: amino acids, purines, pyrimidines, choline, inositol, thiamine, folic acid, biotin, calcium, niacinamide, pyridoxine, riboflavin, thymidine, cyanocobalamin, pyruvate, lipoic acid, magnesium, glucose, trace metals or metals. It may contain salts, and buffers. In some embodiments, the liquid culture medium can contain serum from a mammal. In some embodiments, the liquid culture medium may contain mammalian growth hormone and/or mammalian growth factors. Alternatively, the culture medium may be a minimal medium (eg, a medium containing only inorganic salts, a carbon source, and water). The medium may contain serum from a mammal. The medium may be a chemically defined culture medium in which all chemical components are known.

According to embodiments, the cells are adapted to produce recombinant proteins or peptides, such as immunoglobulins, or other forms of biological products. The term "recombinant" or "engineered" refers to a peptide or polypeptide that is not naturally encoded by endogenous nucleic acids present in an organism (eg, a mammal). Examples of engineered proteins include enzymes, fusion proteins, antibodies, antigen binding proteins, and other types of peptides or proteins used in the pharmaceutical industry or biomedical research.

According to embodiments, the control logic begins to adjust the temperature in the first tank based on the comparison results after a predetermined minimum cell density in the medium of the first tank is measured. The minimum cell density may be, for example, 100000 cells/mL. This can be beneficial because the prediction accuracy is lower at low density values than at high density values. Additionally, the temperature may therefore remain optimal for fast growth rates, especially at the beginning of a cell culture project, and "fine-tuning" of growth rates may be necessary to avoid unnecessary delays in cell culture projects. may be delayed until a later period.

In accordance with some aspects, control logic performs a temperature adjustment of a particular tank only if at least an adjustment time interval has elapsed since a previous temperature adjustment of the particular tank by the control logic.

According to some embodiments, a system allows a user to specify adjustment intervals before a cell culture project begins, and optionally allows a user to dynamically modify adjustment intervals during project execution. For example, the GUI may allow the user to specify target time, target cell density, and additional optimal control parameters, as well as "adjustment intervals."

As used herein, "adjustment interval" means the minimum amount of time that must elapse from a previous temperature adjustment made by the control logic before the control logic can readjust the temperature of the same tank. It's time. Preferentially, the adjustment interval has a duration within the range of approximately 60-120 minutes.

It may be beneficial to use an adjustment interval within the above specified range because the adjustment interval prohibits the control logic from changing the temperature too frequently. This may have had a negative impact on the overall growth rate of the cells, as the cell's metabolism must frequently adapt to different temperatures.

As used herein, the term "production tank" refers to a large-scale tank, particularly a large-scale bioreactor, adapted to culture cells in stationary phase. Production tanks are typically used to produce and optionally recover a desired biological product, such as a protein or peptide. Large tanks have an internal volume of, for example, more than 500 L, 1,000 L, 5,000 L, 10,000 L, 20,000 L, 50,000 L, or 100,000 L. For example, the production tank may be a perfusion bioreactor.

As used herein, the term "perfusion bioreactor" refers to a bioreactor that has an internal volume for culturing a plurality of cells (e.g., recombinant mammalian cells) in a liquid culture, (e.g., an outlet, an inlet, a pump, or other such device), and adding substantially the same volume of replacement liquid culture medium to the bioreactor. refers to a bioreactor that has means (e.g., an outlet, inlet, pump, or other such device) for. Addition of replacement liquid culture medium can be performed substantially simultaneously with, or immediately after, removal of liquid culture medium from the bioreactor. The means for removing liquid culture medium from the bioreactor and the means for adding replacement liquid culture medium may be a single device or system.
[Invention 1001]
A method for controlling proliferation of eukaryotic cells, the method comprising:
- receiving, by control logic (116), a target cell density (114) and a target time (112), the target cell density representing a desired cell density of eukaryotic cells at the target time; Time represents a future point in time; a stage;
- culturing said eukaryotic cells in a culture medium (M1) of a first tank (130);
For each of the multiple time intervals,
- Measuring the cell density (122) of the eukaryotic cells in the culture medium;
- calculating, by the control logic, a predicted cell density at the target time as a function of at least the measured cell density;
- comparing the predicted cell density and the target cell density by the control logic;
・By the control logic,
- by increasing the temperature by 0.1°C to 1°C if the predicted cell density is lower than the target cell density; or
- by lowering the temperature by 0.1°C to 1°C if the predicted cell density is higher than the target cell density;
adjusting the temperature of the culture medium in the first tank if the predicted cell density deviates from the target cell density;
method including.
[Present invention 1002]
The first tank includes an instrument (134) for performing on-line measurements of cell density in the culture medium, and the cell density of eukaryotic cells measured at each of the plurality of time intervals is the on-line measurement. 1001 invention methods.
[Present invention 1003]
-By control logic,
・Achieving the target time;
・The measured cell density is equal to or higher than the target cell density.
automatically terminating at least the temperature adjustment when one of the termination criteria is met;
Any method of the invention, further comprising:
[Present invention 1004]
- providing, by control logic, a graphical user interface (GUI) (110), the GUI allowing the user (118) to enter a target time and target cell density before the control logic begins calculating the predicted cell density; the GUI allows the user to modify the target time and/or the target cell density while the control logic calculates the predicted cell density at multiple time intervals; ,step
Any method of the invention, further comprising:
[Present invention 1005]
- after reaching the target cell density, transferring the culture medium of the first tank and the eukaryotic cells contained therein from said first tank to a second tank (132) which is larger than said first tank;
- receiving, by control logic, a further target cell density and a further target time, the further target cell density representing a desired cell density of the eukaryotic cells at the further target time, the further target time being in the future; A stage representing a point in time;
- culturing the transferred eukaryotic cells in the second tank;
For each of a plurality of further time intervals,
- Measuring the cell density (124) of the eukaryotic cells in the culture medium of the second tank;
- calculating, by the control logic, a further predicted cell density at the further target time, at least as a function of the measured cell density;
- comparing the further predicted cell density and the further target cell density by the control logic;
・By the control logic,
- by increasing the temperature by 0.1°C to 1°C if said further predicted cell density is lower than said further target cell density; or
- by reducing the temperature by 0.1°C to 1°C if said further predicted cell density is higher than said further target cell density;
adjusting the temperature of the culture medium in the second tank if the further predicted cell density deviates from the further target cell density;
Any method of the invention, further comprising:
[Present invention 1006]
a bioreactor, a first tank configured to grow the cell culture as a batch culture, and a second tank larger than the first tank and configured to grow the cell culture as a batch culture; 1005. The method of the invention 1005, wherein the method is used in a seed train culture step to generate a defined minimum number of cells for inoculating a production bioreactor.
[Present invention 1007]
Any method according to the invention, wherein the eukaryotic cell is a mammalian cell, in particular a Chinese hamster ovary (CHO) cell.
[Present invention 1008]
For each of the multiple time intervals,
- defining for the current time interval a current observation interval (138) having a predetermined length and ending at the end of the current time interval;
- receiving all cell densities measured during the current observation interval;
further including;
calculating a predicted cell density as a function of at least the measured cell density comprises extrapolating the cell density measured during the current observation interval up to a target time;
Any method of the present invention.
[Present invention 1009]
1008. The method of the invention 1008, wherein the extrapolation comprises performing a curve fitting operation or regression analysis on the cell density measured during the current observation time.
[Present invention 1010]
A method according to any preceding invention, wherein the plurality of time intervals are of equal length.
[Invention 1011]
- determining whether the current time is close to the target time;
- shortening the length of all future time intervals of the plurality of time intervals in response to determining that the current time is close to the target time;
The method of any one of inventions 1001-1009, further comprising.
[Invention 1012]
The method of any of the inventions 1008-1012, wherein the predetermined length of the observation interval is within the range of 60 minutes to 480 minutes.
[Present invention 1013]
Said book, wherein adjusting the temperature comprises increasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C, or decreasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C. Any method of invention.
[Present invention 1014]
Control logic performs a temperature adjustment of the first tank only if at least a certain adjustment time interval has elapsed since a previous temperature adjustment of the first tank by the control logic, and the duration of the adjustment time interval is between 60 and 60. Any method of the invention, wherein the duration is within 120 minutes.
[Present invention 1015]
A method for providing a desired cell density of eukaryotic cells at a defined future time, the method comprising:
- inputting a desired cell density (114) and a future time (112) by a human user (118) via an interface (110) of the control logic (116), wherein the desired cell density is , representing the desired cell density of eukaryotic cells at the input future time;
- A step of implementing the method for controlling the proliferation of eukaryotic cells according to any one of the present inventions 1001 to 1014, in which the input desired cell density is used as a target cell density; performed such that the input future time is used as a target time and the eukaryotic cells are provided at the target cell density at the target time.
method including.
[Invention 1016]
A system (100) for controlling growth of eukaryotic cells, the control logic comprising control logic (116) and a processor (104) configured to execute the control logic, the control logic operably connected to a first tank (130) containing eukaryotic cells in a liquid (M1), the control logic comprising:
- receiving a target cell density (114) and a target time (112), the target cell density representing a desired cell density of the eukaryotic cells at the target time, and the target time defining a future point in time; to represent;
For each of the multiple time intervals,
- receiving a measured cell density (122) of said eukaryotic cells in said culture medium of said first tank;
- calculating a predicted cell density at a target time as a function of at least said measured cell density;
- comparing the predicted cell density and the target cell density;
・By the control logic,
- by increasing the temperature by 0.1°C to 1°C if the predicted cell density is lower than the target cell density; or
- by lowering the temperature by 0.1°C to 1°C if the predicted cell density is higher than the target cell density;
adjusting the temperature of the culture solution in the first tank when the predicted cell density deviates from the target cell density;
The system (100) that is configured to do so.

Aspects of the invention will now be explained in more detail, by way of example only, with reference to the drawings, in which: FIG.

FIG. 2 is a block diagram of a system for controlling cell proliferation in a first tank and a second tank. Figure 2 shows the forecast time interval and several different current observation intervals. Figure 2 shows the forecast time interval and several different current observation intervals. Figure 2 shows the forecast time interval and several different current observation intervals. Figure 2 shows the forecast time interval and several different current observation intervals. 1 is a flowchart of a method for controlling cell growth by adjusting temperature. 1 shows a system including multiple pre-production tanks and production tanks. Figure 3 illustrates cell density profiles of six cell culture projects. Figure 5 shows in more detail the effect of temperature on cell density signals obtained for one of the six cell culture projects. Figure 5 shows in more detail the effect of temperature on cell density signals obtained for one of the six cell culture projects. Figure 3 shows plots depicting the dependence of online and offline cell density on temperature regulation for three cell culture projects. Further plots representing the dependence of cell density on temperature adjustment are shown for further cell culture projects. Two plots are shown illustrating the ability of the control logic to compensate for oxygen supply failures. Two plots are shown illustrating the ability of the control logic to compensate for oxygen supply failures. Two plots are shown illustrating the ability of the control logic to compensate for pH control failures. Two plots are shown illustrating the ability of the control logic to compensate for pH control failures. FIG. 14 shows three plots illustrating the ability of the control logic to compensate for both oxygen supply and pH control failures. FIG. 14 shows three plots illustrating the ability of the control logic to compensate for both oxygen supply and pH control failures. Two plots are shown illustrating the ability of the control logic to compensate for both oxygen supply and temperature control failures.

DETAILED DESCRIPTION FIG. 1 is a block diagram of a system 102 for controlling cell growth in a first tank 130 and a second tank 132. Growth control for each of the tanks may be performed according to the method illustrated by the flowchart of FIG. 3. Accordingly, the system of FIG. 1 will be described below by also referring to FIG.

Data processing system 102 includes one or more processors 104, main memory 106, and a clock 108. Clock 108 is configured to measure time in absolute or relative terms. It is used to measure prediction time intervals and observation intervals, for example as shown in Figures 2a-2d. Additionally, system 102 includes non-volatile storage medium 114 containing computer-readable instructions that implement control logic 116. The control logic may be a stand-alone application program or a submodule or unit of an application program or software framework used to control one or more bioreactors and/or laboratory equipment.

In addition, storage medium 114 includes computer readable instructions that implement graphical user interface GUI 110. The GUI allows user 118 to specify and enter target time 112 and target cell density 114. Optionally, the GUI additionally allows the user to specify the duration of the observation interval 138 during which cell density is measured and the future in which the controlled cell culture reaches the target cell density 114. Extrapolated to predict time. For example, GUI 110 may be an HTML interface or Java or C# based program logic. In accordance with some embodiments, the GUI 110 not only allows the user to specify target cell density, target time, and observation interval prior to the start of the cell culture project, but also allows the user to specify the target cell density, target time, and observation interval already entered while the cell culture project is in progress. and/or allow the user to modify the stored target cell density, target time, and/or observation interval.

System 102 further includes an interface 120 for receiving measurement data from one or more tanks and/or for sending control commands to one or more tanks. The interface forwards received measurement data to control logic 116 and control commands from the control logic to one or more tanks.

Data processing system 102 may be implemented as a standard computer system, a server computer system, or a mobile communication device, such as a smart phone or tablet computer.

In the example shown in FIG. 1, system 102 is operably coupled to and controls a pre-production bioreactor 130. Bioreactor 130 includes elements 138 for raising or lowering the temperature of the medium. This element may be, for example, a thermostat whose target temperature is controlled by control logic 116. In addition, the pre-production bioreactor 130 includes an online cell densitometer 134 adapted to measure the current cell density in the medium M1 of the first bioreactor 130 without taking a sample. Bioreactor 130 contains a culture medium M1 that does not contain any cells at the time the pre-production cell culture project is initiated.

Before beginning cell growth in pre-production bioreactor 130, user 118 instantiates GUI 110 and computer system 102 to determine a target cell density for a particular cell type and to determine whether the cells within bioreactor 130 are at the target cell density. Enter the target time that defines the time to reach. Typically, the target time is a time at which the particular cell culture project is known to enter a particular growth phase, such as before production or at the end of the growth phase for a production bioreactor, or at the end of a growth phase for a production reactor. Within +/- 20% at the end of the period.

For example, a user may instantiate a GUI at 10.00 am on a Monday. The user wishes to grow pre-production CHO cell cultures, which typically require 48 hours to reach the desired cell density, under "constant temperature" conditions of 37 °C, according to the literature or previous experiments. , one hour later, the pre-production cell culture step can be started in the bioreactor 130 at 11.00 am by inoculating the bioreactor 130 with a thawed CHO cell sample. At 10.00 am, the user enters the target time 112, target cell density, and observation time interval. For example, the user may enter 4 million cells/ml for the target cell density and "Monday 11.00am + 48 hours" as the future target time. In addition, the user can set the prediction time interval to, for example, about 1 minute and the observation interval to, for example, 90 minutes. Users can even set the target time to "Monday 11.00am + 47 hours" to reduce the time required to perform the cell culture step by one hour.

After configuring the control logic via the GUI 110, at 11.00 am the user starts pre-production cell culture by inoculating the cell-free medium M1 in the bioreactor 130 with a thawed CHO cell sample. The starting temperature of the tank may be 37°C.

In a first step 302 of the control methods described herein, e.g., at approximately 11.00 a.m. on Monday, the control logic determines from the storage medium 114 the target cell density 114, the target time 112, and Optionally read the duration of observation interval 138. The loaded target cell density represents the desired cell density of eukaryotic cells at the entered target time (Monday 11.00 am + 48 hours). It will be appreciated that this time may be specified in a variety of different formats depending on the embodiment, for example, as an absolute time or as a relative time starting from the current time or starting from a particular future time.

At step 304, eukaryotic cells are cultured in culture medium M1 in first tank 130. For example, step 304 may include inoculating medium M1 in tank 130 with CHO cells. According to other embodiments, it is also possible to perform step 304 immediately before step 302, and to perform both steps 302, 304 essentially in parallel.

The control logic may then automatically identify a series of "predicted time intervals" I ₁ , ..., I ₄₃₆₇ , also referred to as "time intervals", of a predetermined length, such as one minute. The first time interval may begin, for example, just before or after the time of inoculation.

For each of the plurality of time intervals 306 (eg, at the beginning or end of each of the time intervals), the following sub-methods are performed.

At step 308, the online cell densitometer 134 measures the cell density 122 of eukaryotic cells in the culture medium M1 in the first tank 130. Measurements 122 are sent to control logic 116 via interface 122.

In response to receiving the measured cell density value 122, the control logic calculates, at step 310, the predicted cell density at the target time "Monday 11.00 am + 48 hours" as a function of at least the currently measured cell density. do. In particular, cell density at a target time can be predicted as a function of multiple cell densities measured during sliding observation intervals as shown in Figure 2.

Additionally, the control logic compares the predicted cell density and the target cell density at step 312.

At step 314, control logic determines whether the predicted cell density is higher than, lower than, or the same as the target cell density. If the control logic determines that the predicted cell density is the same as the target cell density entered by the user, or at least within an acceptable threshold band around the target cell density, then The current temperature is maintained at stage 318, either actively or passively. If the control logic determines that the predicted cell density is less than the target cell density entered by the user or less than the lower tolerance threshold below the target cell density, the control logic causes the thermostat to A temperature adjustment control command 126 is sent to the thermostat 138 that causes the current temperature of the culture medium within the reactor 130 to be increased by, for example, 0.4°C. Alternatively, if the control logic determines that the predicted cell density is greater than the target cell density entered by the user or greater than an upper tolerance threshold above the target cell density, the control logic controls the thermostat in step 316. sends a temperature adjustment control command 126 to thermostat 138 that causes the current temperature of the medium within bioreactor 130 to decrease, for example, by 0.4°C. In accordance with embodiments, a GUI allows one 104 for a user 118 to specify the amount of temperature change to be imposed on the culture medium for each predicted time interval in which a temperature adjustment is deemed necessary.

The sub-method comprising steps 308-318 above may be performed completely automatically for each of a plurality of time intervals of relatively short duration, typically one to several minutes.

In accordance with some aspects, a sub-method, such as step 314, may include a further check to determine whether termination criteria have been met. The control logic determines that termination criteria have been met, for example, if a predetermined maximum number of predicted time intervals has elapsed, a target cell density has been reached, or a predetermined maximum duration of the cell culture project has been exceeded. do.

The method described above, including the sub-methods 306 to 318 that are carried out repeatedly, is suitable for starting a cell culture project in which cells are grown in a pre-production cell culture bioreactor 130 and subsequently carried out in a second tank 132, e.g. a production bioreactor. can be used to grow cells to a desired density. The production bioreactor also includes a thermostat 140 that is controlled by the control logic via respective control commands 128, and an online cell density meter 136 that presents measured cell density values to the control logic 116. Measured cell density and temperature control commands 128 are communicated to and from the control logic via interface 120. The cell culture medium M1 in the second tank may have basically the same composition as the culture medium in the first tank, or may have a different composition. A second cell culture project in a second tank is initiated by transferring the cells and optionally also the medium contained in the first tank to the second tank after reaching the target cell density. Additional cell-free medium is added to the second tank, which causes dilution of the transferred cells in the second tank.

Prior to growing cells in the second tank 132, the user 118 reconfigures the target time 112 according to the characteristics of the production cell culture process in the second tank. For example, the first cell culture project may have ended exactly at the target time, i.e. Monday 11.00am + 48 hours, i.e. Wednesday 11.00am. Transfer of cells from the first tank to the second tank, reconfiguration of control logic, and replenishment of further medium may require 30 minutes. It is known from the literature that a cell culture project in which cells are grown at 37 °C in a target tank starting from the cell number provided by the first tank takes 74 hours. Therefore, on Wednesday at 11:15, the user reconfigures the control logic by entering a new target time "Wednesday 11:30 + 74 hours" or "Saturday 1:30 p.m." be able to. After moving the cells from the first tank to the second tank, the second cell culture project will begin at 11:30 a.m. on Wednesday, and the second cell culture project will start in the second tank 132 as described above for the first cell culture project. Cell growth is controlled by adjusting the temperature of the medium.

Figure 2a shows a plurality of "predicted time intervals" I ₁ ,..., I ₄₃₆₇ , also referred to as "time intervals", starting at the time the cell culture is inoculated with eukaryotic cells. For each time interval, for example at the end of each time interval, the current cell density in the medium in the control bioreactor is measured. For example, at the end of the first time interval I ₁ the cell density CD _M1 is measured. At the end of the second time interval _I2 , the cell density CD _M2 is measured. At the end of the third time interval _I3 , the cell density CD _M3 is measured. The same applies hereafter. In addition, for each time interval, or at least for a time interval starting after the observation interval has elapsed, the predicted cell density C _P is calculated by extrapolating the cell density measurements obtained during the current observation interval. .

For example, by extrapolating the cell densities CD _M1 , CD _M2 , ..., CD _M5 , CD _M6 measured during the observation time interval at the end of time interval I ₆ , the predicted cell density CD at target time 112 _P6 is predicted. In addition, the control logic compares the predicted cell density CD _P6 and the target cell density 114, and at the end of time interval I ₆ , controls the temperature of the control bioreactor if the predicted cell density CD _P6 is significantly different from the target cell density 114. adjust.

Figure 2b illustrates the data processing steps performed by the control logic at the end of time interval _I7 . By extrapolating the cell densities CD _M2 , CD _M3 , ..., CD _M6 , CD _M7 measured during the observation time interval, the cell density CD _P7 at the target time 112 is predicted. In addition, the control logic compares the predicted cell density CD _P7 and the target cell density 114, and at the end of time interval I ₇ , if the predicted cell density CD _P7 is significantly different from the target cell density 114, the control bioreactor temperature Adjust.

FIG. 2c illustrates the data processing steps performed by the control logic at the end of time interval I ₁₁ . By extrapolating the cell densities CD _M6 , CD _M7 , ..., CD _M10 , CD _M11 measured during the observation time interval, the cell density CD _P11 at the target time 112 is predicted. In addition, the control logic compares the predicted cell density CD _P11 and the target cell density 114, and at the end of time interval I ₁₁ , if the predicted cell density CD _P11 is significantly different from the target cell density 114, the control bioreactor temperature Adjust.

FIG. 2d illustrates the data processing steps performed by the control logic at the end of time interval I ₄₃₆₅ . By extrapolating the cell densities CD _M4358 ,..., CD _M4365 measured during the observation time interval, the cell density CD _P4365 at the target time 112 is predicted. In addition, the control logic compares the predicted cell density CD _P4365 and the target cell density 114, and at the end of time interval I ₄₃₆₅ , if the predicted cell density CD _P4365 is significantly different from the target cell density 114, the control bioreactor temperature Adjust. Compared to the time intervals shown in Figures 2a-2c, the time interval shown in Figure 2d is significantly shorter. For example, the time intervals shown in Figures 2a-2c may have a duration of several minutes, for example 15 minutes, and the time intervals shown in the right part of Figure 2d may have a duration of 1 minute. For example, the control logic may automatically use a shorter time interval when determining that the current time is close to the target time, eg, 2-3 hours before the target time. This can be beneficial since at this stage of the project the cell density is already high and minor measurement errors can have a significant impact on the predicted target time. By using shorter time intervals, the robustness of the method to anomalous measurements may be improved, and thus the number of predictions and prediction accuracy may be increased.

FIG. 4 shows a system that includes a plurality of pre-production tanks 400, 412, 130 and a production tank 132. Pre-production tanks 400, 412, 130 are used in a seed train process to provide a sufficient number of cells to start a production cell culture project within production bioreactor 132. The thawed cell line sample 400 is inoculated into the first pre-production bioreactor, and the first pre-production cell culture is started in the first tank 410. When the first target cell concentration is reached in the first tank, the culture medium and cells in the first tank are moved to the second pre-production tank 412, and a second pre-production cell culture is started in the second tank 412. When the second target cell concentration is reached in the second tank, the culture medium and cells in the second tank are moved to the third pre-production tank 130, and a third pre-production cell culture is started in the third tank 130. When the third target cell concentration is reached in the third tank, the culture medium and cells in the third tank are moved to the production tank 132, and production cell culture is started in the production tank 132. At least the production bioreactor may include a gas inlet line or pipe. For example, a single gas inlet line or pipe can be used to deliver ambient (or already expanded) compressed air from a special source to the bioreactor. The ambient air or compressed air may consist of a mixture of gases, in particular N2, O2 and CO2, which are typical of the Earth's atmosphere or have a different composition. Additionally or alternatively, a single gas inlet line or pipe, or any other gas inlet line or pipe, may be used to deliver individual gases such as N2, O2, and CO2 to the bioreactor. can be used. For example, the bioreactor may include a microsparger to generate finely dispersed gas bubbles from the incoming gas to improve aeration of cells within the tank.

Tanks 410, 412, 130, 132 each include a thermostat 138, 406, 408, 140 and an on-line cell density meter 134, 402, 404, 136 and are operably coupled to control logic 116, thereby providing temperature control. receive. GUI 110 allows user 118 to specify target times and target cell densities for each of the four cell culture projects individually. For example, the GUI may be used to set the target time and target cell density for the first tank 410, as well as any further process parameters such as interval length, observation interval length, amount of temperature adjustment per time interval, etc. May include window W410. The GUI displays a second window W412 for setting the above parameter values for the second tank 412, a third window W414 for setting the above parameter values for the third tank 130, and a third window W414 for setting the above parameter values for the production tank 132. may include a fourth window W132 for.

Figure 5 shows six different batch cell culture projects carried out to test the effect of temperature regulation on the proliferation rate of eukaryotic cells. In each of the cell culture projects described in Figure 5ff, CHO cell lines were grown in a 2 liter bioreactor B-DCU (Sartorius). An EVO200 system (Fogale, now Hamilton) was used to perform online cell density measurements. Culture of cells for all batches was performed in shake flasks according to reference cell culture protocols. The bioreactor used for each project was equipped with a pO2 electrode (Clark type) and a pH gel electrode. The working volume was always 1.3 liters and the bioreactor was vented (headspace venting and sparring). A constant flow of 50 ccm of carbon dioxide-enriched nitrogen was passed through the bioreactor through the headspace. Pure oxygen was introduced via spurring according to the pO2 regulator. Gas flow was regulated via a mass flow controller (MFC). All online measurements were recorded via MFCS. A Fogale EVO200 system was used for online cell counting. The dielectric constant (dielectric conductivity) of the culture was measured and converted to a cell density value by a constant factor. These values are referred to as "online" cell densities. EVO200 system measurements were read through specially programmed software and input into control logic configured to adapt the bioreactor temperature.

To check whether the dielectric constant-based cell densities accurately reflected the actual cell densities, they were sometimes compared with offline cell concentrations measured by analyzing bioreactor culture fluid samples. For offline controls of cultures, samples were initially taken at least once a day and cell density was measured on a Cedex cell analyzer. During the later stages of the project, daily offline cell density measurements were partially omitted. Although the dielectric constant is not constant throughout the fermentation process and depends on several parameters, it was nevertheless possible to work with only one conversion value within the relevant cell concentration range.

Two of the six batches were reference batches grown at constant temperature throughout the cell culture project. In the remaining four batches, the effect of temperature on cell growth was investigated by lowering the temperature during day intervals according to different schedules. Each of the two reference batch cultures and the four test batch cultures were always started at a standard temperature of 36.5° C. approximately every other day to condition the cells to a high growth rate. Temperature changes were then performed in four test batches 3-6 as shown in the figure: batch 3 was grown at 36.5°C for 9.3 days and then at 35°C; batch 4 was grown at 36.5°C for 12 days. Batch 5 was grown for 15 days at 36.5°C and then grown at 33°C; batch 6 was grown at 36.5 The cells were grown for 18 days at 0.degree. C., then 1 day at 35.degree. C., then 1 day at 34.degree. C., and the next day and all subsequent days at 32.degree.

For each of the batches, a continuous smooth line 500 indicates the cell density measured by converting the dielectric constant measure to a "measured cell density" by a single experimentally determined correction factor. The stepped curve 502 shows off-line measurements taken from time to time to ensure that the dielectric constant accurately reflects the actual cell density over the entire temperature range tested. To perform offline control measurements of cell density, samples were taken and cell density was measured using a Cedex cell analyzer.

Figure 5 shows that cell growth rates can be controlled by temperature regulation in existing bioreactor systems and that the dielectric constant scale accurately reflects cell density over the relevant temperature range. Although the temperature of the cell culture medium was modified during the cell culture process, the produced proteins and peptides examined so far did not show any deviation from the biological products of the reference cell culture.

Figures 6 and 7 show in more detail the effect of temperature on cell density signals obtained for two of the six cell culture projects. In both figures, a stepwise decrease in temperature 504 results in a decrease in growth rate, and the measured permittivity multiplied by a single constant conversion factor is determined by occasional offline, sample-based cell density measurements 500 As confirmed, it was a reliable measure of cell density 502 (i.e., the number of cells in a given medium volume).

A further cell culture project (“batch”) was performed to assess the effect of pH changes on growth rates (not shown). Additional batches revealed that over a wide pH range from pH 6.75 to pH 7.1, no short-term effects of the modified pH could be observed. The effects of pH changes, if any, appeared only with long waiting times. Starting from pH 6.75 and converting to pH 6.8, the effect was observed after about 1.5 days, at pH 6.7 after about 1 day, and at pH 6.6 after half a day (see batch 4). If the pH was lowered below 6.75, the culture required considerable time to regenerate after the new pH rose to the standard value (6.9-7.1). Therefore, contrary to temperature-based growth control, pH-based growth control suggests that in the physiological pH range, pH changes do not seem to have a significant and rapid effect on cell growth, and that pH outside the physiological range When used, it has been observed that the cells face the problem of requiring additional time to recover from extreme pH values, thereby slowing down the cell culture project and reducing its flexibility. Ta. Cell culture growth rates were found to be more strongly dependent on temperature changes in the physiological temperature range of about 32-38° C. than on pH changes in the physiological pH range of about 6.8-7.2. Furthermore, cell recovery is significantly worse from non-physiological pH values than from abnormally low temperatures. Therefore, it was observed that pH adjustment is not suitable for flexibly controlling cell culture growth rates.

Therefore, in accordance with embodiments, temperature was used to provide automatic control of cell culture projects because even small temperature changes have significant, rapidly reversible effects on growth behavior.

According to an embodiment, the control logic calculates the biomass value from the dielectric constant value measured by an online (dielectric conductivity-based) measurement device (e.g. EVO200) or from the dielectric constant value by multiplying the dielectric constant by a conversion factor. (cell density value) to calculate the current proliferation rate.

The conversion factor is determined by performing several offline cell density measurements in parallel with the permittivity measurements and then converting one or more reference cells by identifying a conversion factor that returns the offline cell density when multiplied by the permittivity value. Determined in cultivation project.

According to embodiments, the time interval for performing permittivity offline measurements and predicting cell density at the target time is 60 seconds. The dielectric constant values obtained for the intervals are logarithmized (base e) by control logic and stored temporarily or permanently in a storage medium. Optionally, offline cell densities are measured from time to time and stored in relation to each online cell density. Logarithmized cell density values are subjected to linear regression over a predetermined observation interval (eg, 90 minutes). After each new dielectric constant (and thus online cell density) measurement (eg, after each 60 seconds), a new calculation is made. The slope of the regression line is then used to determine the growth rate μ of the cells over the current observation interval. This value μ naturally rises and falls for some measurements, but is fixed within a specified observation interval. The longer the observation interval, the more stable the calculated proliferation rate and the more accurate the predicted cell density at the target time.

Using the calculated growth rate, a prediction of the expected cell density at the target time is calculated, and the predicted cell density is compared to the target cell density set by the user. The target time and target cell density are defined at the beginning of the batch process, which may also be the point of initiation of the control logic. If the prediction results in a cell density above the target cell density, the temperature is decreased by a small amount (e.g., 0.4 degrees); if the predicted cell density is below the target cell density, the temperature is increased slightly ( For example, 0.2 degrees). After the temperature change, a new observation interval is automatically started and the cycle begins again. This continues until the target time is reached or another termination criterion is met.

For example, an externally controllable water bath (equipped with countercooling) can be used to regulate the temperature of the bioreactor. However, any control system that allows to transmit external values (set points, control parameters) to elements that can raise and lower the temperature of the medium in the tank can be used.

Figure 8 shows further text culture projects (batches 6, 7, and 8) carried out as described for the test batches shown in Figures 5, 6, and 7 online (e.g., dielectric constant-based or Based on online microscopy) cell density 802, 812, 822, occasionally measured offline cell density values (symbols) 800, 810, 820, and temperature signals (dashed lines) 804, 814, 824 are shown. Online and offline cell density values are plotted on a logarithmic scale and temperature on a linear scale. The vertical arrow indicates the target time to reach the target cell density, eg 4,000,000 cells/ml (horizontal line). Figure 8 shows that each target time density was reached very precisely at the desired target time.

Batches 6, 7, and 8 essentially have the same starting conditions, the same target cell density, but different target times. Each target was hit almost exactly. The temperature profile of the batch, especially batch 8, is such that the control logic also takes into account the current growth rate of the cells and, if appropriate, the different growth phases (growth phase, stationary phase or transition phase after inoculation). Show what you get.

Figure 9 shows the online cell densities 902, 912, 922 of further text culture projects (batches 9 and 10) carried out as described for the test batches shown in Figures 5, 6 and 7, and occasionally measured offline. Cell density values (symbols) 900, 910, 920 and temperature signals (dashed lines) 904, 914, 924 are shown. The plots were made as described for FIG. 8 and also include the values of batch 8 (with reference numbers 900, 902, 904, different from 820, 822, 824) as a reference. Batch 9 was a cell culture project designed to replicate the target time and target cell density (3 days long) of Batch 8, but from the temperature of the plot, at the time of inoculation, Batch 8 and Batch It is clear that the 9 media had different temperatures. Nevertheless, the program logic is able to provide target cell densities at target times with great accuracy for both Batch 8 and Batch 9.

Batch 10 started with a similar target cell density and starting temperature as Batch 9, but the target time was different (4 days). The results of Batch 10 can show that the same target cell density could be achieved with the same cell density as Batch 9, supplemented for a full day. Embodiments of the invention therefore provide a means to clearly define the end point of fermentation within a time window of at least one day, e.g. allow for regulation.

The above test was performed using the cell line T074OPT B048WCB. Further tests were performed based on cell cultures of the CHO cell line T072. For further test cell culture projects, the duration of the conditioning interval (60-120 minutes), the starting cell density, the target time (approximately 3-4 days, or 67-95 hours), and the height of the temperature change (0.2-0.5 °C ) was slightly modified. In all cell lines, target cell density could be reached within the target time. All temperature curves were observed to be different from each other. Therefore, the growth control system individually and successfully controlled each cell culture. In the event of a large temperature adjustment, e.g. 0.5-1.0°C, the length of the adjustment interval was increased (preferentially to at least 90 min) so that the cells had more time to adapt to the new temperature. .

FIGS. 10-15 show that even though a cell culture project may encounter one or more technical problems and failures during the project, temperature-regulated growth control performed in accordance with embodiments of the present invention can achieve the desired results at the target time. demonstrated that it is possible to provide cell densities of .

Technical failures can seriously jeopardize the success of cell culture projects. Corrective measures are typically performed manually as soon as possible. Typically, poor driving can be remedied within 4 hours. However, delays in cell growth caused by technical failures often result in an extension of the duration of whole cell culture projects.

In a series of tests, we evaluated the ability of the control logic to compensate for delays in cell growth caused by typical technical failures by simulating the failure of one or more regulatory controls of the bioreactor. Failure scenarios tested included oxygen supply failure, oxygen electrode failure, temperature control failure, pH control (CO2 supply) failure, and others. In each of the tests, one or more control functionalities (e.g., O2 partial pressure regulation, CO2 partial pressure regulation, etc.) were stopped for 4 hours, or operating parameters (e.g., pH) were changed for approximately 4 hours. It was set to an undesirable value. Control functionality was then resumed or operating parameters reset to desired or typical values. The control logic, particularly target time and target cell density settings, were not changed during the study.

The results of some of the tests are presented in the form of plots in Figures 10-15. The cell cultures used in the studies shown in Figures 10-15 were grown as described for the cell cultures mentioned in the figure legend of Figure 5. The numbering of the cell cultures ("batches") used in the test now starts from the beginning. Therefore, cell cultures 4-9 in Figures 10-15 are not identical to the cell cultures whose data form the basis of Figures 5-9.

Figures 10 and 11 each show two plots illustrating the ability of the control logic to compensate for oxygen supply failures. Testing of the ability of the control logic to compensate for failures in the oxygen supply was performed by stopping operation of the oxygen regulator during the time interval 48h to 52h (Figure 10) or during the time interval 55h to 59h (Figure 11) . Figures 10 and 11 clearly show that without oxygen (oxygen signal) the cells do not proliferate further (upper plot, decrease in the slope of the cell density curve). The control logic determines the decrease in growth rate and, given the low growth rate, predicts that the cell density at the target time will be below the target cell density and increases the temperature with a 1-2 hour delay (Figures 10 and 11 bottom plot). Despite the temporary failure of oxygen supply, it was possible to achieve the desired cell density at the set target time.

Figures 12 and 13 each show two plots illustrating the ability of the control logic to compensate for pH control failures and resulting undesirable pH values. A test of the ability of the control logic to compensate for a pH control failure is to rapidly raise the pH control set point to an undesired value during the time interval 48h to 52h (Figure 12) or during the time interval 72h to 76h (Figure 13). This was done by changing to . Figures 12 and 13 show that pH changes do not significantly affect cell proliferation and, as a result, the control logic compensates for the pH-induced proliferation effects without any problems and increases the target cell density at the target time. Indicates that the

FIG. 14 shows three plots illustrating the ability of the control logic to compensate for both oxygen supply and pH control failures. Again, the control logic could provide the target cell density at the target time.

FIG. 15 shows two plots illustrating the ability of the control logic to compensate for both oxygen supply and temperature control failures. 26 hours after inoculation, failure of oxygen supply was simulated for approximately 4 hours. The system responded very strongly, as expected, to compensate for the decrease in cell proliferation. The control logic reacted to the drop in oxygen supply by increasing the temperature during the time of failure. After approximately 12 hours, temperature control failure was simulated by turning off the thermostat. Within about 4 hours, the reactor temperature dropped to a value of 30-31°C. During this time, the control logic continues to predict future cell densities and compares the calculated density to the target cell density, but is unable to successfully communicate and enforce each new temperature set point. After the temperature controller was switched on again, the temperature controller received a new temperature set point from the control logic. This results in a slight overflow of the actual temperature, which is then rapidly compensated. Cell cultures were grown in a controlled manner for the remainder of the batch run time. Again, the control logic could provide the target cell density at the target time.

Therefore, it has been successfully demonstrated that temperature-based growth regulation according to embodiments of the present invention is robust to and can compensate for various technical problems and failures. Intermittent failures of oxygen or pH control are tolerated without problems and their effects on growth are compensated. Even temperature control failures can still be mitigated.

Claims (16)

A method for controlling proliferation of eukaryotic cells, the method comprising:
- receiving, by control logic (116), a target cell density (114) and a target time (112), the target cell density representing a desired cell density of eukaryotic cells at the target time; Time represents a future point in time; a stage;
- culturing said eukaryotic cells in a culture medium (M1) of a first tank (130);
For each of the multiple time intervals,
- Measuring the cell density (122) of the eukaryotic cells in the culture medium;
- calculating, by the control logic, a predicted cell density at the target time as a function of at least the measured cell density;
- comparing the predicted cell density and the target cell density by the control logic;
・By the control logic,
- by increasing the temperature by 0.1°C to 1°C if the predicted cell density is lower than the target cell density; or - by increasing the temperature by 0.1°C to 1°C if the predicted cell density is higher than the target cell density. By lowering the temperature,
A method comprising adjusting the temperature of the culture medium in the first tank if the predicted cell density deviates from the target cell density. The first tank comprises an instrument (134) for performing an on-line measurement of cell density in a culture medium, and the cell density of eukaryotic cells measured at each of the plurality of time intervals is the on-line measurement. The method described in Section 1. -By control logic,
・Achieving the target time;
- the measured cell density is equal to or higher than the target cell density . Method described in 2 . - providing, by control logic, a graphical user interface (GUI) (110), the GUI allowing the user (118) to enter a target time and target cell density before the control logic begins calculating the predicted cell density; the GUI allows the user to modify the target time and/or the target cell density while the control logic calculates the predicted cell density at multiple time intervals; 4. The method of any one of claims 1-3 , further comprising the steps . - after reaching the target cell density, transferring the culture medium of the first tank and the eukaryotic cells contained therein from said first tank to a second tank (132) which is larger than said first tank;
- receiving, by control logic, a further target cell density and a further target time, the further target cell density representing a desired cell density of the eukaryotic cells at the further target time, the further target time being in the future; A stage representing a point in time;
- culturing the transferred eukaryotic cells in the second tank;
For each of a plurality of further time intervals,
- Measuring the cell density (124) of the eukaryotic cells in the culture medium of the second tank;
- calculating, by the control logic, a further predicted cell density at the further target time, at least as a function of the measured cell density;
- comparing the further predicted cell density and the further target cell density by the control logic;
・By the control logic,
- by increasing the temperature by 0.1°C to 1°C if said further predicted cell density is lower than said further target cell density, or - by increasing the temperature if said further predicted cell density is higher than said further target cell density. By lowering the temperature by 0.1℃ to 1℃,
The method of any one of claims 1 to 4 , further comprising adjusting the temperature of the culture medium in the second tank if the further predicted cell density deviates from the further target cell density. . a bioreactor, a first tank configured to grow the cell culture as a batch culture, and a second tank larger than the first tank and configured to grow the cell culture as a batch culture; 6. The method of claim 5, wherein the method is used in a seed train culture step to generate a defined minimum number of cells for inoculating a production bioreactor. 7. A method according to any one of claims 1 to 6 , wherein the eukaryotic cell is a mammalian cell, in particular a Chinese hamster ovary (CHO) cell. For each of the multiple time intervals,
- defining for the current time interval a current observation interval (138) having a predetermined length and ending at the end of the current time interval;
- further comprising receiving all cell densities measured during the current observation interval;
calculating a predicted cell density as a function of at least the measured cell density comprises extrapolating the cell density measured during the current observation interval up to a target time;
The method according to any one of claims 1 to 7 . 9. The method of claim 8, wherein extrapolating comprises performing a curve fitting operation or regression analysis on the cell density measured during the current observation time. 10. A method according to any one of claims 1 to 9 , wherein the plurality of time intervals are of equal length. - determining whether the current time is close to the target time;
- in response to determining that the current time is close to the target time, further comprising: - shortening the length of all future time intervals of the plurality of time intervals. The method described in section. 12. A method according to any one of claims 8 to 11 , wherein the predetermined length of the observation interval is within the range of 60 minutes to 480 minutes. Claim wherein the step of adjusting the temperature comprises increasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C, or decreasing the temperature by 0.1°C to 0.5°C, preferentially 0.1°C to 0.2°C. The method described in any one of 1 to 12 . Control logic performs a temperature adjustment of the first tank only if at least a certain adjustment time interval has elapsed since a previous temperature adjustment of the first tank by the control logic, and the duration of the adjustment time interval is between 60 and 60. 14. A method according to any one of claims 1 to 13 , within the range of 120 minutes. A method for providing a desired cell density of eukaryotic cells at a defined future time, the method comprising:
- inputting a desired cell density (114) and a future time (112) by a human user (118) via an interface (110) of the control logic (116), wherein the desired cell density is , representing the desired cell density of eukaryotic cells at the input future time;
- implementing the method for controlling the proliferation of eukaryotic cells according to any one of claims 1 to 14, wherein the proliferation control is performed when the input desired cell density is the target cell density; The method is performed such that the future time used and input is used as a target time and the eukaryotic cells are provided at the target cell density at the target time. A system (100) for controlling the growth of eukaryotic cells, comprising control logic (116) and a processor (104) configured to execute the control logic, the control logic operably connected to a first tank (130) containing eukaryotic cells in a liquid (M1), the control logic comprising:
- receiving a target cell density (114) and a target time (112), the target cell density representing a desired cell density of the eukaryotic cells at the target time, and the target time defining a future time point; to represent;
For each of the multiple time intervals,
- receiving a measured cell density (122) of said eukaryotic cells in said culture medium of said first tank;
- calculating a predicted cell density at a target time as a function of at least said measured cell density;
- comparing the predicted cell density and the target cell density;
・By the control logic,
- by increasing the temperature by 0.1°C to 1°C if the predicted cell density is lower than the target cell density; or - by increasing the temperature by 0.1°C to 1°C if the predicted cell density is higher than the target cell density. By lowering the temperature,
A system (100) configured to adjust the temperature of the culture medium in the first tank if the predicted cell density deviates from the target cell density.

Images